Metraipolis › archive › dc_feb3c2e5343b

Skaling: Chinchilla's Exponents Meet Kaplan's Coupling — arXiv:2608.07222

Source
arXiv (en)
Original
https://arxiv.org/abs/2608.07222
License (as archived)
CC0-1.0
Archived
2026-10-19T09:46:57.313Z
SHA-256 of the archived text
5e90756b64d43a67de366045a4c7d8a3a1e226c0350a2de8f22e4a6094122494
Stable URL of this page
https://metraipolis.com/documento/dc_feb3c2e5343b

How to cite

Skaling: Chinchilla's Exponents Meet Kaplan's Coupling — arXiv:2608.07222 — arXiv (en), https://arxiv.org/abs/2608.07222 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_feb3c2e5343b (Metraipolis archive dc_feb3c2e5343b, sha256 5e90756b64d43a67de366045a4c7d8a3a1e226c0350a2de8f22e4a6094122494)
BibTeX
@misc{dcfeb3c2e5343b,
  title = {Skaling: Chinchilla's Exponents Meet Kaplan's Coupling — arXiv:2608.07222},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2608.07222},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_feb3c2e5343b; sha256 5e90756b64d43a67de366045a4c7d8a3a1e226c0350a2de8f22e4a6094122494},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2608.07222. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Neural scaling laws are foundational for language model development, yet standard formulations systematically under- and overestimate loss at data-scarce and overtraining extremes. This failure originates in the underlying assumption that model size and training data impact the loss independently. To address this, we introduce the Skaling law, a generalized functional form that couples model capacity and data through a single interaction exponent. This simple extension reduces the Mean Absolute Percentage Error (MAPE) by 1.5-3x across both interpolation and extrapolation regimes. When paired with a sparse grid strategy restricted to low-compute regimes, the Skaling law achieves accurate full-grid extrapolation using approximately 10x less compute than uniform sweeps. By enabling reliable performance prediction from small-scale experiments, the Skaling law provides a more robust and resource-efficient framework for allocating compute budgets in next-generation model training.

Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-08-10 in cs.CL).
Categorie arXiv: Computation and Language (cs.CL).
Pagina del lavoro: https://arxiv.org/abs/2608.07222 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito).
Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina.
Letto il 2026-10-19T09:46:37.239Z dalla pagina https://arxiv.org/catchup/cs.CL/2026-08-10?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).